TL;DR
This paper presents findings from two airborne campaigns (CAATER-1 and CAATER-2) over Western Europe, using high-precision instruments to map CO2 variability in the lower atmosphere. The study highlights how airborne measurements, combined with back-trajectory analysis and CO co-measurements, can effectively distinguish between fossil fuel emissions and biospheric CO2 uptake, even at regional scales.
What Problem Does This Paper Address?
Accurately quantifying regional carbon sources and sinks is crucial for verifying climate change mitigation efforts. However, understanding the spatial and temporal variability of CO2, especially within the continental planetary boundary layer (PBL), remains a significant challenge. Traditional ground-based monitoring stations often struggle to capture the full picture due to their limited spatial coverage and potential local influences. Inverse modeling, a common technique for estimating fluxes, relies heavily on accurate atmospheric transport models and representative in-situ measurements, which are often lacking over continents. This paper aims to address this gap by using airborne measurements to provide a more detailed, regional-scale view of CO2 distribution and its drivers.
Methods
The study utilized two intensive airborne campaigns, CAATER-1 (May 2001) and CAATER-2 (October 2002), aboard a Falcon 20 jet aircraft. The aircraft was equipped with high-precision instruments for continuous in-situ measurements of CO2, CO, and Radon-222. Flask samples were also collected for cross-validation. A key methodological aspect was the use of HYSPLIT-4 back-trajectory analysis to determine the origin of air masses sampled during the flights. Additionally, the study leveraged the correlation between CO and CO2 concentrations (ΔCO/ΔCO2 ratios) to help differentiate between fossil fuel emissions and other CO2 sources, particularly in the CAATER-2 campaign where CO data was available. The researchers also compared their airborne data with ground-based station measurements to assess the representativeness of different station types.
Datasets
- CAATER-1 and CAATER-2 Airborne Campaign Data: High-precision in-situ measurements of CO2, CO, and Radon-222, along with flask samples.
- HYSPLIT-4 Back-trajectory Data: Used to trace the origin of air masses.
- NOAA/NCEP Global Reanalysis Data: Meteorological fields used to drive HYSPLIT-4.
- ORCHIDEE Ecosystem Model Data: Used for Net Ecosystem Exchange (NEE) flux maps (mentioned as a source for Figure 1, with further details in a companion paper).
- Fossil Fuel (FF) Emission Maps: From Andres et al. (1996).
- Oceanic Flux Data: From Takahashi et al. (1999).
- SPOT VGT-4 Satellite Data: Used for Normalized Difference Vegetation Index (NDVI) to assess photosynthetic activity.
- Ground Station Data: From various sites across Europe (e.g., Cabauw, Monte Cimone, Mace Head, Puy-de-Dôme, Schauinsland) for comparison with airborne measurements.
Key Insights and Ideas
- Airborne Campaigns as Powerful Tools: The study strongly reinforces the utility of airborne platforms for capturing regional CO2 variability over short periods, providing a more comprehensive picture than fixed ground stations alone.
- Distinguishing Emission Sources: The combination of CO2, CO measurements, and back-trajectory analysis proved effective in identifying and characterizing air masses influenced by anthropogenic emissions (fossil fuel combustion) versus those affected by biospheric processes (photosynthesis).
- CO:CO2 Ratios as Fingerprints: The ΔCO/ΔCO2 ratios, when analyzed for anthropogenic-influenced air masses, showed good agreement with national emission inventories, suggesting this approach can help pinpoint the origin of fossil fuel emissions even when air masses have traveled for several days.
- Vertical Gradients Matter: Significant differences in CO2 concentrations were observed between the planetary boundary layer (PBL) and the free troposphere (FT), with much higher variability in the PBL. This vertical structure is critical for understanding regional carbon budgets.
- Ground Station Representativeness: The study confirmed that low-level ground stations are representative of local scales, while mountain-top stations can represent regional scales in the FT, provided they are not influenced by valley air.
Key Results
- Mean CO2 Concentrations: CAATER-1: 370.1 ± 4.0 ppm; CAATER-2: 371.7 ± 5.0 ppm.
- PBL vs. FT Variability: CO2 variability in the PBL was 5 to 8 times higher than in the FT.
- PBL standard deviation: 4.0 ppm (CAATER-1), 5.7 ppm (CAATER-2).
- FT standard deviation: 0.5 ppm (CAATER-1), 1.1 ppm (CAATER-2).
- PBL-FT CO2 Jump: 3.7 ppm (CAATER-1), -0.3 ppm (CAATER-2).
- Regional CO2 Depletion/Enrichment: Observed CO2 levels ranged from as low as 355 ppm over rural areas (biospheric uptake) to around 380 ppm over urbanized/industrial regions (anthropogenic emissions).
- ΔCO/ΔCO2 Analysis: For anthropogenic-influenced air masses, R² values ranged from 0.33 to 0.88, with slopes (ΔCO/ΔCO2) from 2.42 to 10.37 ppb CO ppm⁻¹ CO2, showing good agreement with national inventories.
- Zonal Gradient (CAATER-2): A striking ~11 ppm CO2 gradient was observed in the mid-PBL, with higher concentrations in Western Europe than in the East.
Limitations and Future Work
- Limited Campaign Duration: The campaigns were relatively short (4 days each), representing specific seasons. The authors suggest that more airborne campaigns are needed to assess if the observed zonal gradient in autumn is a consistent phenomenon or a punctual event.
- Model-Flux Discrepancies: The study notes that model fluxes did not always align with observed biospheric sinks, particularly in the mid-PBL, suggesting that biospheric fluxes are not always accurately modeled.
- Future Work: The authors emphasize the need for more airborne vertical profiles to better understand regional gradients. They also suggest that future studies could benefit from incorporating tracers like CO, Radon-222, and carbon isotopes to further differentiate the roles of anthropogenic, biospheric, and oceanic sources and sinks. A companion paper (Part 2) delves into model-observation comparisons and flux calculations.
Summary updated